What is OpenStreetMap?

OpenStreetMap (OSM) has contributed to the shift in perception of who can map and how it can be done (Haklay and Weber 2008). Conceived in 2004 by Steve Coast, the project aims to create a free and editable map that everyone could access and use. The project’s focus on making its data accessible to local communities is one of the reasons behind OSM being community-driven with an emphasis on local knowledge in mapping. Putting community and local knowledge at the heart has many benefits, such as:

  • keeping data up-to-date;
  • empowering citizens to represent their local environments;
  • encouraging the participation of citizens in policy making;

The importance of local knowledge is also highlighted in the LTN 1/20 guide in the context of ensuring successful implementation of a scheme. Hence, OSM can be a tool to foster a bottom-up approach in active travel planning through the inclusion of citizens in the data generation process. This, consequently, can be used to vocalise their needs leading to an increased likelihood of achieving higher uptake levels.

What is the structure of OpenStreetMap data?

OSM relies on a simple data structure and flexible tagging to describe geographical objects. Indeed, OSM deliberately chose not to adapt the existing standards for geographic information (Haklay and Weber 2008) as the goal was to simplify the use and maintenance of a project.

The structure of OSM data is composed of three key elements (also called entities or primitives):

  • nodes
  • ways
  • relations

In general, OSM can be described as a graph with edges and vertices that are connected or isolated (Bennett 2010). Tags (key=“value” pairs) are used to describe their geographical features.

Nodes

Nodes are points in space that are defined by latitude and longitude. They can be used on their own or as part of a way or a relation.

An example of a point in active transport context would be a bicycle rental.

Ways

Ways are lines formed by linking at least two nodes (i.e., representing two ends), however the node can belong to more than one way. They always have a direction, represented by the arrows in the editor, even if it is not meaningful (e.g., a wall). These kinds of ways can be defined as open ways as they have different start and end nodes.

Ways can also be closed if the start and end nodes are identical. This is usually used to represent areas, or polygons, but also roundabouts or closed barriers, such as a closed wall.

A park itself is a good example of an area.

Relations

Relation is a list of nodes, ways, and/or relations. It is used to model geographic relationships between objects. These relations should not contain only closely connected elements, having no more than 300 members (elements) per relation.

Most likely the most important relation in the context of active transport is route but there are also other relations, such as public transport schemes and restrictions.

Tags

Tags are not exactly data structures but they play a vital role in OSM because they define geographic features. In other words, without tags one would not know if a node represents a bicycle rental.

A tag is a concept that refers to a key=“value” pair. A key describes a topic while value is a specification of a yopic. For instance, highway=“cycleway” indicates that it is a highway (i.e., a line) and, specifically, a cycleway. It is important to note that a value is not always unique in a sense that it can also be a key. For example, there is a “cycleway” tag, which is used to further detail cycling infrastructure. Hence, it is important to pay attention to the syntax of a tag to understand its meaning and use in a given context.

Moreover, each OSM element can have multiple tags and the use of tags is not limited to certain data elements. Thus, theoretically, a point could be highway="footway". The lack of restrictions is inherent in OSM. It relies on the mappers to make decisions that are the most sensible in their local contexts. This does not mean, however, that there are no conventions about the most appropriate way of tagging something. A simple example would be a depreciation of a sidewalk="none" tag which should be changed to sidewalk="no". A more intricate example is a question concerning the mapping of sidewalks as currently there are two proposed schemes: one considering a sidewalk as a tag on a highway and another proposing to map footways as separate ways.

Figure 1 shows 8 different tags that are important in transport research and their presence/absence in central Leeds.


The most important tag for transport research is highway as it indicates the type of a road. Table 1 shows a list of highway values to work with. Table 2 shows a list of 8 tags which are useful to get started.

Table 1. Values of highway=*
Highway Meaning
cycleway Pats for cycling
footway Footpaths
living_street Streets where pedestrians have priority over cars
motorway Motorways or freeways
motorway_link Motorways or freeways
path Unspecified paths
pedestrian Pedestrian only streets
primary Primary roads
primary_link Primary roads
residential Roads in residential areas
road Roads in residential areas
secondary Secondary roads, typically regional
service Service roads for access to buildings, parking lots, gas stations, etc.
steps Flights of steps on footpaths
tertiary Tertiary roads, typically local
tertiary_link Tertiary roads, typically local
track For agricultural use
trunk Important roads, typically divided
trunk_link Important roads, typically divided
unclassified Smaller local roads
Note:
This table has been adapted from Chan and Cooper (2019).

Table 1.
Tag Meaning
highway It indicates the road type.
cycleway Used for mapping cycling infrastructure.
foot Provides information on the legal access for pedestrians.
maxspeed Maximum speed limit for a particular road, railway, or waterway.
kerb Indicates the presence (or absence) and height of a kerb.
wheelchair Indicates places and ways suitable for wheelchair users.
width Actual width of a way.
lit Indicates presence of lighting.
Note:
Links to relevant OSM wiki pages:
1 highway: https://wiki.openstreetmap.org/wiki/Key:highway
2 cycleway: https://wiki.openstreetmap.org/wiki/Key:cycleway
3 foot: https://wiki.openstreetmap.org/wiki/Key:foot
4 maxspeed: https://wiki.openstreetmap.org/wiki/Key:maxspeed
5 kerb: https://wiki.openstreetmap.org/wiki/Key:kerb
6 wheelchair: https://wiki.openstreetmap.org/wiki/Key:wheelchair
7 width: https://wiki.openstreetmap.org/wiki/Key:width
8 lit: https://wiki.openstreetmap.org/wiki/Key:lit

OSM in transport research

In 2017, Barrington-Leigh and Ball argued that over 80% of all roads are mapped in OSM and in most of the European and North American countries it is more than 95%. While the extensiveness of the road network is impressive, the paper focused on roads intended for vehicle circulation. While it can be assumed that some of the roads can and are used by, for instance, cyclists but it does not reveal anything about the cycling infrastructure which is needed to promote active modes of travel (Handy, van Wee, and Kroesen 2014). Active travel, however, brings a variety of benefits ranking from mental and physical health to environmental benefits (Parkin 2018). Furthermore, it will be even more important in the post-pandemic world due to the reduced public transport capacities. Thus, to encourage citizens not to return to private motor transports, it is important to ensure that infrastructure for active travel meets their travel needs.

To address the lack of knowledge about the quality of cycling infrastructure in OSM, (ferster_etal_2020?) conducted a study in which the data of OSM was compared to official open data from city government in 6 Canadian cities. They found that there was a high level of concordance between the datasets, but that consistent labeling (i.e., topology) of cycling infrastructure in OSM is a challenge that needs to be considered. Importantly, however, it is noted that some of the inconsistency might have resulted from the fact that some types of cycling infrastructure is new in Canada, such as cycling tracks. Although it shows that it might take some time for OSM mappers to (re)map their surroundings, it also calls attention to the importance of communicating new infrastructural developments so that different types can be easily identified, mapped, and tagged.

In general, there are examples of OSM being used to map both cycling and pedestrian networks. For example, Orozco et al. (2020) developed an algorithm to improve cycling network by connecting disconnected segments. They argued that for a highly segmented city like London (over 3000 segments compared to Copenhagen’s 300) it will take great investments to improve connectivity. Kasemsuppakorn and Karimi (2013) examined OSM for pedestrian network construction, Graser (2016) used OSM to evaluate different approaches to integrating open spaces in pedestrian routing systems while Novack, Wang, and Zipf (2018) developed a customized pleasant pedestrian routing system.

OSM has also been joined to other datasets, such as Strava or GPS and travel diary data to improve the understanding of cyclists route choice in Glasgow (Alattar, Cottrill, and Beecroft 2021) and North England (Yeboah and Alvanides 2015), respectively. Differently from cycling, however, pedestrian networks appear to be examined more in relation to accessibility. Recently Cohen and Dalyot (2021) used OSM to develop a wayfinding algorithm for planning accessible and safe routes to blind pedestrians. Boularouk, Josselin, and Altman (2017) argued that the use of OSM could help to develop cheaper assistive technologies for people with disabilities who, compared to people without disabilities, have lower median incomes in the UK (Francis-Devine 2021).

Finally, OSM data has been used in a number of open-access, and for social good in general, projects:

  • A/B Street is a simulation game that demonstrates how various small changes in road infrastructure affects its users.
  • Cyclestreets is a UK-wide route planning system for cyclists.
  • WheelMap provides a map that shows wheelchair (in)accessible places.

Why use OSM?

So far, we have outlined what is OSM and how it was used in transport research. However, what kind of benefits would the adoption of open data yield?

The first one, already mentioned, is the potential of OSM to reduce costs of the tools and systems that rely on geospatial data (Boularouk, Josselin, and Altman 2017). Second, OSM could be used for conducting comparative analysis between different countries and/or regions, especially in the absence of national geographic dataset (Nelson et al. 2021). Third, the use of OSM data could increase the transparency, reproducibility, and replicability of (geospatial) research and, thus, its robustness (Brunsdon and Comber 2021). Moreover, opening spatial analyses in general, and the data used in particular, would aid in unveiling the “black box” algorithms that, in the age of Big Data, constitute a part in the decision-making processes in the daily lives of citizens (Brunsdon and Comber 2021). Fourth, it fosters citizen science and citizen participation in democratic processes by empowering citizens to engage in (local) geographic knowledge production (Peters 2020). Given that OSM mappers tend to be highly engaged with their local communities (Budhathoki and Haythornthwaite 2013), adoption of OSM could lead to increased accountability and promotion of evidence-based decision making in transport research (Lovelace 2021). This, consequently, could lead to an increased uptake in cycling, walking, and wheeling as a preferred modal choice.

The role of OpenInfra in transport research

The adoption of OSM brings a number of benefits to transport research, however it might not always be clear on how to use, or even download, OSM data. OpenInfra aims to ease the use of OSM in transport research for those who would benefit from it the most - citizens and policy makers. The emphasis is also added on inclusivity and accessibility; not only is the goal to improve the accessibility of OSM data but also demonstrate how it can be used to plan inclusive infrastructures for cycling, walking, and wheeling. To achieve this, the package contains a number of functions to make OSM data processing easier as well as hands-on articles, or vignettes, to get started.

Bibliography

Alattar, Mohammad Anwar, Caitlin Cottrill, and Mark Beecroft. 2021. “Modelling Cyclists’ Route Choice Using Strava and OSMnx: A Case Study of the City of Glasgow.” Transportation Research Interdisciplinary Perspectives 9 (March): 100301. https://doi.org/10.1016/j.trip.2021.100301.
Barrington-Leigh, Christopher, and Adam Millard-Ball. 2017. “The World’s User-Generated Road Map Is More Than 80% Complete.” PLOS ONE 12 (8): e0180698. https://doi.org/10.1371/journal.pone.0180698.
Bennett, Jonathan. 2010. OpenStreetMap. Packt Publishing Ltd.
Boularouk, Said, Didier Josselin, and Eitan Altman. 2017. “Ontology for a Voice Transcription of OpenStreetMap Data: The Case of Space Apprehension by Visually Impaired Persons.” In World Academy of Science, Engineering and Technology. London, United Kingdom.
Brunsdon, Chris, and Alexis Comber. 2021. “Opening Practice: Supporting Reproducibility and Critical Spatial Data Science.” Journal of Geographical Systems 23 (4): 477–96. https://doi.org/10.1007/s10109-020-00334-2.
Budhathoki, Nama R., and Caroline Haythornthwaite. 2013. “Motivation for Open Collaboration: Crowd and Community Models and the Case of OpenStreetMap.” American Behavioral Scientist 57 (5): 548–75. https://doi.org/10.1177/0002764212469364.
Cohen, Achituv, and Sagi Dalyot. 2021. “Route Planning for Blind Pedestrians Using OpenStreetMap.” Environment and Planning B: Urban Analytics and City Science 48 (6): 1511–26. https://doi.org/10.1177/2399808320933907.
Ferster, Colin, Jaimy Fischer, Kevin Manaugh, Trisalyn Nelson, and Meghan Winters. 2020. “Using OpenStreetMap to Inventory Bicycle Infrastructure: A Comparison with Open Data from Cities.” International Journal of Sustainable Transportation 14 (1): 64–73. https://doi.org/10.1080/15568318.2018.1519746.
Francis-Devine, Brigid. 2021. “Income Inequality in the UK.” House of Commons Library Briefing Paper.
Haklay, Mordechai, and Patrick Weber. 2008. OpenStreetMap: User-Generated Street Maps.” IEEE Pervasive Computing 7 (4): 12–18. https://doi.org/10.1109/MPRV.2008.80.
Handy, Susan, Bert van Wee, and Maarten Kroesen. 2014. “Promoting Cycling for Transport: Research Needs and Challenges.” Transport Reviews 34 (1): 4–24. https://doi.org/10.1080/01441647.2013.860204.
Kasemsuppakorn, Piyawan, and Hassan A. Karimi. 2013. “A Pedestrian Network Construction Algorithm Based on Multiple GPS Traces.” Transportation Research Part C: Emerging Technologies 26 (January): 285–300. https://doi.org/10.1016/j.trc.2012.09.007.
Lovelace, Robin. 2021. “Open Source Tools for Geographic Analysis in Transport Planning.” Journal of Geographical Systems 23 (4): 547–78. https://doi.org/10.1007/s10109-020-00342-2.
Nelson, Trisalyn, Colin Ferster, Karen Laberee, Daniel Fuller, and Meghan Winters. 2021. “Crowdsourced Data for Bicycling Research and Practice.” Transport Reviews 41 (1): 97–114. https://doi.org/10.1080/01441647.2020.1806943.
Novack, Tessio, Zhiyong Wang, and Alexander Zipf. 2018. “A System for Generating Customized Pleasant Pedestrian Routes Based on OpenStreetMap Data.” Sensors 18 (11): 3794. https://doi.org/10.3390/s18113794.
Orozco, Luis Guillermo Natera, Federico Battiston, Gerardo Iñiguez, and Michael Szell. 2020. “Data-Driven Strategies for Optimal Bicycle Network Growth.” Royal Society Open Science 7 (12): 201130. https://doi.org/10.1098/rsos.201130.
Parkin, John. 2018. Designing for Cycle Traffic: International Principles and Practice. London: ICE Publishing.
Peters, Michael A. 2020. “Citizen Science and Ecological Democracy in the Global Science Regime: The Need for Openness and Participation.” Educational Philosophy and Theory 52 (3): 221–26. https://doi.org/10.1080/00131857.2019.1584148.
Yeboah, Godwin, and Seraphim Alvanides. 2015. “Route Choice Analysis of Urban Cycling Behaviors Using OpenStreetMap: Evidence from a British Urban Environment.” In OpenStreetMap in GIScience: Experiences, Research, and Applications, edited by Jamal Jokar Arsanjani, Alexander Zipf, Peter Mooney, and Marco Helbich, 189–210. Lecture Notes in Geoinformation and Cartography. Cham: Springer International Publishing. https://doi.org/10.1007/978-3-319-14280-7_10.